Medical Imaging with Distortion Correction

1. A method of generating an image, the method comprising: obtaining distorted image data of a subject brain, the distorted image data comprising a time series of three-dimensional image tensors generated at least in part from an echo planar imaging session of the subject brain, wherein distortion of the distorted image data is caused by external factors shifting locations of individual image data; deriving a derived three-dimensional tensor from the distorted image data; determining a non-rigid alignment function to non-rigidly align the derived three-dimensional tensor to a reference tensor producing a non-rigidly aligned derived 3D tensor; determining a rigid alignment function to rigidly align the non-rigidly aligned derived 3D tensor to the reference tensor; and producing distortion-corrected image data by applying the rigid alignment function and the non-rigid alignment function to the time series of three-dimensional image tensors.

2. The method of claim 1 , wherein determining the rigid alignment function comprises: selecting a portion of the derived three-dimensional tensor as an anchor; finding a target-portion of the reference tensor that corresponds to the anchor; and finding a translation of the anchor that minimizes a different function between the anchor and the target portion.

3. The method of claim 2 , wherein selecting a portion of the derived three-dimensional tensor as an anchor comprises finding a center of the derived three-dimensional tensor.

4. The method of claim 1 , wherein determining the non-rigid alignment function comprises determining an affine transformation that specifies a difference in shape between the derived three-dimensional tensor to the reference tensor.

5. The method of claim 4 , wherein the transformation, when applied to the distorted image data, transforms a first portion of the distorted image data and does not transform a second portion of the distorted image data.

6. The method of claim 1 , wherein deriving a derived three-dimensional tensor from the distorted image data comprises averaging voxel values of each address location across a plurality of volumetric images of the distorted image data.

7. The method of claim 1 , the method further comprising one of the group consisting of i) displaying the distortion-corrected image data to a user; ii) storing the distortion-corrected image data; and iii) performing clinical analysis for the subject brain using the distortion-corrected image to produce a result different than what would be produced performing the clinical analysis on the distorted image data.

8. The method of claim 1 , wherein the echo planar imaging session of the subject brain is free of a sensing for generation of a field gradient map and wherein the producing of the distortion-corrected image data is performed free of a field gradient map.

9. The method of claim 1 , wherein the echo planar imaging is one of the group consisting of functional magnetic resonance (fMRI) and imaging and diffusion tractography (DT).

10. The method of claim 1 , wherein the reference tensor represents a T1 image.

11. A system for generating an image, the system comprising: one or more computers and one or more storage devices on which are stored instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising: obtaining distorted image data of a subject brain, the distorted image data comprising a time series of three-dimensional image tensors generated at least in part from an echo planar imaging session of the subject brain, wherein distortion of the distorted image data is caused by external factors shifting locations of individual image data; deriving a derived three-dimensional tensor from the distorted image data; determining a non-rigid alignment function to non-rigidly align the derived three-dimensional tensor to a reference tensor producing a non-rigidly aligned derived 3D tensor; determining a rigid alignment function to rigidly align the non-rigidly aligned derived 3D tensor to the reference tensor; and producing distortion-corrected image data by applying the rigid alignment function and the non-rigid alignment function to the time series of three-dimensional image tensors.

12. The system of claim 11 , wherein determining the rigid alignment function comprises: selecting a portion of the derived three-dimensional tensor as an anchor; finding a target-portion of the reference tensor that corresponds to the anchor; and finding a translation of the anchor that minimizes a different function between the anchor and the target portion.

13. The system of claim 12 , wherein selecting a portion of the derived three-dimensional tensor as an anchor comprises finding a center of the derived three-dimensional tensor.

14. The system of claim 11 , wherein determining the non-rigid alignment function comprises determining an affine transformation that specifies a difference in shape between the derived three-dimensional tensor to the reference tensor.

15. The system of claim 14 , wherein the transformation, when applied to the distorted image data, transforms a first portion of the distorted image data and does not transform a second portion of the distorted image data.

16. The system of claim 11 , wherein deriving a derived three-dimensional tensor from the distorted image data comprises averaging voxel values of each address location across a plurality of volumetric images of the distorted image data.

17. The system of claim 11 , the operations further comprising one of the group consisting of i) displaying the distortion-corrected image data to a user; ii) storing the distortion-corrected image data; and iii) performing clinical analysis for the subject brain using the distortion-corrected image to produce a result different than what would be produced performing the clinical analysis on the distorted image data.

18. The system of claim 11 , wherein the echo planar imaging session of the subject brain is free of a sensing for generation of a field gradient map and where the producing of the image data is performed free of a field gradient map.

19. The system of claim 11 , wherein the echo planar imaging is one of the group consisting of functional magnetic resonance (fMRI) and imaging and diffusion tractography (DT).

20. The system of claim 11 , wherein the reference tensor represents a T1 image.

21. The method of claim 1 , wherein the distorted image data contains shifted locations of intensity values without changes to those intensity values.

22. The method of claim 1 , wherein air in a sinus cavity of the subject influences the echo planar imaging session resulted in the distortion of the distorted image data.

23. The system of claim 11 , wherein the distorted image data contains shifted locations of intensity values without changes to those intensity values.

24. The system of claim 11 , wherein air in a sinus cavity of the subject influences the echo planar imaging session resulted in the distortion of the distorted image data.